Performance monitoring system for the bt_api ecosystem. Provides real-time metrics collection, exchange health monitoring, and integrations with Prometheus, Grafana, and ELK stack.

• MetricsCollector — Centralized metrics collection with global collector support
• Counter / Gauge / Histogram — Standard metric types with Prometheus naming conventions
• PerformanceTimer — High-precision timing utilities for performance profiling

• @monitor_performance — Decorator for automatic performance monitoring
• @monitor_async_performance — Async-aware performance monitoring decorator
• @monitor_calls — Call count and frequency monitoring

• ExchangeHealthMonitor — Health check framework for exchange connections
• HealthCheck / HealthCheckResult — Standardized health check types
• HealthStatus — Enum for UP / DOWN / DEGRADED states

• PrometheusExporter — Push metrics to Prometheus Pushgateway
• GrafanaDashboardBuilder — Generate Grafana dashboard JSON configs
• ELKIntegration — Elasticsearch / Logstash / Kibana integration via ElasticsearchClient and LogstashHandler

• SystemMetricsCollector — CPU, memory, disk, network usage tracking
• BusinessMetricsCollector — Custom business-level metric collection

pip install bt_api_monitor

The canonical distribution and import are now bt_api_monitor. The former bt_api_monitoring import remains as a deprecated compatibility shim for one migration window; update imports before the next breaking release.

• Python >= 3.9
• psutil >= 5.9.0
• aiohttp >= 3.9.0
• pydantic >= 2.0.0

from bt_api_monitor import MetricsCollector, counter, gauge, histogram

# Use the global collector
collector = MetricsCollector()

# Increment a counter
counter("requests_total", tags={"endpoint": "/api/v1/ticker"}).inc()

# Set a gauge
gauge("active_connections", tags={"exchange": "binance"}).set(42)

# Record a histogram
histogram("request_latency_seconds", tags={"method": "GET"}).observe(0.123)

# Start global monitoring
from bt_api_monitor import start_global_monitoring, stop_global_monitoring
start_global_monitoring()
# ... your code ...
stop_global_monitoring()

from bt_api_monitor import ExchangeHealthMonitor, HealthCheckFactory

monitor = ExchangeHealthMonitor()
check = HealthCheckFactory.create("binance_spot", timeout=5.0)
result = monitor.check_sync(check)
print(f"Status: {result.status}, latency: {result.latency_ms}ms")

from bt_api_monitor import start_prometheus_exporter, stop_prometheus_exporter

start_prometheus_exporter(port=9090)
# Metrics are now available at http://localhost:9090/metrics

Full documentation available at bt_api_py documentation.

MIT License